h a l f b a k e r y
Normal isn't your first language, is it?

meta:

account: browse anonymously, or get an account and write.

 user: pass:
register,

# Donut Piston

cream filled.
 (+4) [vote for, against]

If you were to use a ball-on-a-stick as a piston you'd notice that, no matter how much you move the stick around, the ball still retains contact with the cylinder walls as a circular line.

This idea is to lop off the top and bottom of the ball, leaving just enough to cope with the angular movement of the connecting rod (stick), eliminating both wristpin and piston-guide skirt, leaving a very lightweight one-piece piston/conrod.

A compression "collar" fitted to the curvature of the piston wall provides the functionality of compression rings. Unlike compression rings which are very tightly expanded against the cylinder wall, the compression collar is only as tight as it needs to be at any given time. Passages lead from the piston crown to the gap between piston wall and collar, so while the collar by itself is just sprung tightly enough to effectively seal against the -14psi of the intake stroke, the higher pressures of the other strokes push it against the cylinder wall proportionally to the pressure inside the combustion chamber.
---

Note that "donut" is just a convenience. Obviously the shape of the piston wall surface is defined by mounting an imaginary disk on top of the conrod and seeing what shape it traces as the conrod moves. If the piston is allowed to spin (to even out wear from side loading) then it's a true spheric equatorial section. If not then the necessary structure is reduced to a "bow tie" cross-sectional surface formed by wobbling the imaginary disk only in one plane.

For example, a piston with a 3.5" bore, 4" stroke with a 6" conrod would require about 37º of movement which would make the piston wall about an inch wide at its widest points (along the plane of conrod travel).

... no reason, I just thought it was neat.

 — FlyingToaster, Sep 02 2011

[2 fries shy of a happy meal, Sep 02 2011]

Mesmerizing http://www.pjrobert.../foursphericons.php
[2 fries shy of a happy meal, Sep 02 2011]

Parker Hannafin http://www.parker.c...0MOTORS&Wtky=MOTORS
patented 30 years ago for use in bent-axis hydraulic pump/motors [FlyingToaster, Sep 05 2011]

here's one http://thekneeslide...-piston-two-stroke/
I think it's thicker than it needs to be. [FlyingToaster, Oct 04 2012]

Cool visual but wouldn't it wear out very quickly with so little contact area?
 — 2 fries shy of a happy meal, Sep 02 2011

 It actually has quite a bit of contact area... just not all at the same time: the piston wobbles along with the conrod on its way up and down (though at TDC and BDC it's flat) so the contact circle is spread out across the curvature of the side.

 On some engines the piston-head and/or cylinder is tilted a little to spread out the side-loading from the compression and combustion stroke more evenly.

 The piston could also be allowed to spin freely on top of the conrod to even wear.

 (As a technical aside, an equatorial spheric section as the post describes, lends itself to the freely spinning piston.

If the piston and conrod are truly one solid piece, the top and bottom of the piston are more accurately shaped as elliptical rather than circular; since the conrod (and thus the piston) only moves angularly in one plane, the further away from that plane the side of the piston is, the less curvy it has to be, in fact the cross-section would/could be a bow-tie)
 — FlyingToaster, Sep 02 2011

hmmm, I've seen something like this somewhere...
 — 2 fries shy of a happy meal, Sep 02 2011

I found a pump that (looks like it) might use this idea: the piston's basically a ball... but it uses regular looking compression rings which doesn't make sense since the fuel would just slide between the rings at high angular displacement. [edit: took another look, see my next anno]
 — FlyingToaster, Sep 02 2011

 I think I found what it was your idea reminded me of, [link] 1. It's nothing close but well worth a look still... and if you want a chuckle, google sphericon piston.

and [link] 2 was just too cool not to post too
 — 2 fries shy of a happy meal, Sep 02 2011

 Ah here we go, Parker-Hannafin uses the same principle in their line of hydraulic "elbow" pump-motors: two turning plates angled to each other, hinged pistons extruding from one, poking into cylinders on the other... there's not much angular play though.

... and [2f] I always get a kick out of that.
 — FlyingToaster, Sep 02 2011

 This would be good for a pump, but not so good for an internal combustion engine.

 The problem is going to be lubrication, and sealing.

 Most conventional cylindrical pistons have 2, 3, sometimes 4 rings (1 & 2 compression, oil control, scraper).

 The spheroid is going to have a single point of circumferential contact which will change constantly; "rings" per se are therefore impracticable.

 Also, the rings accommodate thermal expansion in the piston and block.

 For a ceramic piston and block, however, this might work well.

 A further idea would be to make the big end not in the form of a journal bearing but as a captive balljoint, This would allow the piston and conrod to rotate about its principal axis, equalising wear.

 A 2-stroke design with Schnurle porting and a truly hemispherical head would be an attractive concept.

[+] for innovative thinking, even if of doubtful practicality.
 — 8th of 7, Sep 02 2011

 Instead of compression rings, a compression collar which follows the piston wall curvature; see paragraph 3, got the seal covered as well (rather nicely if I do say so myself).

 Tentatively, the collar fits into 2 ring grooves, one at top the other at the bottom. Pressure is equalized between the combustion chamber and the tiny gap between piston wall and collar via small holes in the piston crown. So during the compression and combustion stroke the section of collar that happens to be crank-side at the time is forced outwards. During the exhaust stroke not so much, and during the intake stroke not at all, but the collar is pre-sprung to be able to seal against minus one atmosphere.

Lubrication I wasn't going to bother with posting, but one idea is if we make it a squishy piston crown then oil can be forced out of the piston onto the walls when the pressure of combustion hits it (oil feed via the conrod).
 — FlyingToaster, Sep 02 2011

Of course the easier solution that uses existing parts would simply be to have a whole bunch of compression rings in an x shape.
 — FlyingToaster, Sep 03 2011

Couldn't the entire combustion chamber be squishy and stretchy, then no need for moving parts or sliding seals?
 — pocmloc, Sep 03 2011

 well... "squishy" as in "distorts elastically at 300psi" sort of thing.

 //entire combustion chamber... stretchy// it is, that's what the piston's for.

I'm gonna stick with the "collar" idea though... Pistons are sized such that when the engine is as hot as it's going to get, they're really snug and the compression rings don't poke out of their gaps (or at least that's what I think happens), this way it would be the piston+collar which would be sized like that, when it's cooler there's a bit of a gap in between them, but the collar is still sprung outwards, and pressed outwards even more during compression and combustion.
 — FlyingToaster, Sep 03 2011

 // or at least that's what I think happens //







 The piston is sized such that when the engine is at its correct operating temperature (and new or reconditioned), there is a very small gap between the piston and the cylinder wall.

 This includes the skirt. Over time, the bore and the piston both wear asymmetrically, causing an effect called "piston slap".

 The compression rings provide the wiping seal between the piston and the cylinder bore. They seal against the bore on their outer circumference; they seal against the piston on their upper or lower face, depending on whether they are on the induction or compression stroke. There is for all practical purposes no seal on the inner face of "spring" piston rings.

 There are exceptions to this, for example the solid bronze rings used in some early Diesel engines, but these require a two=part piston.

 The compression rings always "poke out" of the machinings in the piston, as they are manufactured to be oversize for the bore by a very small amount. The circumferential force is fairly constant during all piston movements.



[FLYINGTOASTER], ARE YOU PAYING ATTENTION ?
 — 8th of 7, Sep 03 2011



 hmm... feel free to substitute ". With this idea" for ", this way" in my previous annotation :D



//Please pay attention. You may be asked questions later.// I may be too much of a hick to know whether that should be a colon or a semicolon, but it sure ain't the right place for a period.
 — FlyingToaster, Sep 03 2011

Given that the text is clearly presented as verbal dialogue //"the placement of punctuation is clearly at the caprice of the author to control Tone and Pacing. The reader should deliver the words as if they represent complete sentences even if they are fragments, commas semi-colons and colons each receiving due pause, and periods indicating a full stop. The first word after a full stop should be emphasized even if it is a fragment."// Written dialogue has rules, but the form has no real constraints.
 — WcW, Sep 03 2011

 But the second sentence is clearly connected to the first.

 "Please pay attention", why ? because "you may be asked questions later".

Perfect place for a full colon if you ask me.
 — FlyingToaster, Sep 03 2011

 //piston slap//

 Side-loading on the collar is spread out over quite an area (as it moves). Not so much the cylinder though. You can always angle the cylinder a bit to even it out between sides.

 But the thing that's concentrating the sideload is also the thing that's reducing friction.

 //Schneurle// I don't see any advantage here; a bit of a disadvantage in that the ports will be exposed and hidden rather slowly as the piston's curved wall passes by...

[edited out a long explanation of why the wobbling will mitigate cylinder wear... which it doesn't. We need a post concerning the relationship between halfbaker and petard]
 — FlyingToaster, Sep 04 2011

 // exposed and hidden rather slowly as the piston's curved wall passes by //

 Shirley not. The port will start to be exposed as soon as the "point of obturation" passes it; there's very little difference between this and a flat-headed cylindrical piston, if you look closely at the geometry, and the way the piston curves away from the cylinder wall.

 Go on, we dare you.

 // the relationship between halfbaker and petard //

We design 'em, we make 'em, you get hoist by 'em ...
 — 8th of 7, Sep 04 2011

 Unless you were making the point of superior gas flow vector from the curve, but...

 Given the posted example (normal'ish engine component dimensions/geometry), the piston-collar is an inch wide; at BDC there's half an inch between the contact circle and the top of the piston.

Which means that the gap between the top edge of the piston and the cylinder wall is 0.07 inches. That's not much difference from a 'squared edge' piston.
 — FlyingToaster, Sep 04 2011

However I think that [8/7] and [2fries] are right: the reason it's not in an IC engine is because of sideload. <link> to Parker Hannafin who holds some sort of patent on a "spherical piston" and engineering cutaway drawings of several of their bent-axis hydraulic pump/motors<link> show the same thing, for the same purpose, except they're using compression rings instead of a collar which I think is rather unimaginative.
 — FlyingToaster, Sep 05 2011

Has this donut been piston?
 — rcarty, Sep 05 2011

Not really: that's mostly just banana-cream filling.
 — FlyingToaster, Sep 08 2011

ewwww you guys!
 — xandram, Sep 08 2011

The perfect place to point out the latest link which shows a working model of the main bits of this idea: an actual IC engine, not just the PF hydraulic pump.
 — FlyingToaster, Oct 05 2012

 [annotate]

back: main index